Internet addiction and depressive symptoms: a dose-response effect mediated by levels of physical activity

Abstract Introduction There are several negative impacts on the physical and mental health of people presenting internet addiction, including development of mood disorders, such as depression. Objective The aim of the study was to evaluate the association between internet addiction and depressive symptoms, as well as to test the mediating role of physical activity in this association. Methods A cross-sectional study was conducted with undergraduate students from three universities (one private and two public institutions) in southern Brazil. Depressive symptoms were measured with the Beck Depression Inventory (BDI-21), internet addiction with the Internet Addiction Test (IAT-20), and physical activity with the International Physical Activity Questionnaire (IPAQ – short version). Poisson regression and the Karlson-Holm-Breen mediation method were used for statistical analyses. Results We observed a dose-response relationship between internet addiction and depressive symptoms. Levels of physical activity mediated the association between moderate internet addiction and depressive symptoms, accounting for 10.7% of the effect observed. Conclusion Internet addiction can be detrimental to individuals’ health, contributing to development of depressive symptoms. Along with psychological and pharmacological therapies, prescription of physical activities is recommended.


Introduction
Since its creation in the late 1960s, the internet has been gaining more and more space in our lives, allowing us to connect more easily and for longer than at any other time in human history. Recently, due to wireless technologies, and with the popularization of smartphones, access to the internet has become easier, changing the way we communicate and spread information, making the internet a ubiquitous part of everyday life and contemporary culture. 1 It is a tool used by individuals of all ages; but younger people are more likely to use the internet with all its purposes and functionalities. 2 It is indisputable that the internet has catalyzed development of society and its impacts echo in different contexts, such as education, health, politics, and communication, especially due to the exponential increase in the capacity to store and transmit information. 3 However, this "symbiotic" relationship with the internet can also have negative effects. One example is internet addiction, which, despite not being recognized as a mental disorder in diagnostic manuals, has similar consequences to gambling and alcohol and drug abuse disorders. 4 One example of this similarity is the neuroanatomical and neurochemical changes in the brain's reward circuitry resulting from excessive internet use. 5 The international literature has documented important negative impacts on the physical and mental health of people with internet addiction, including development of mood disorders, such as depression. [5][6][7] Undergraduates seem to be a group especially susceptible to development of internet addiction, since studies indicate high prevalence of this outcome, ranging from 16.7% 8 to 48.5%. 9 Depression also seems to be highly frequent in this subgroup, considering that results from metaanalyses indicate that 28.5% of Brazilian 10 and 30.6% of US university students 11 have significant depressive symptomatology, proportions almost three-fold higher than among the general populations of both countries. 11,12 Moreover, evidence from distinct cultural contexts indicates that the association between internet addiction and depression is very consistent among undergraduate students. 6,7,9,13 However, the mechanisms that explain this association are less clear.    Adjusted analysis controlled for sex, age, socioeconomic level, skin color, university, alcohol consumption, smoking, physical activity, and body mass index (see Table 2). Sample of 1,026 undergraduate students, Pelotas, Brazil, 2017.  Figure 2 -A) Global, direct, and indirect effects (coefficients, odds ratios, and p-values) between mild internet addiction and depressive symptoms, mediated by levels of physical activity. B) Global, direct, and indirect effects (coefficients, odds ratios, and p-values) between moderate internet addiction and depressive symptoms, mediated by levels of physical activity. Analysis adjusted by variables with p < 0.2 for their associations with depression (i.e., university type, alcohol consumption, and smoking; see Table 2). Sample of 1,026 undergraduate students, Pelotas, Brazil, 2017.

Discussion
There are several ways to classify pathological use of the internet, which makes it difficult to compare different investigations. 4,26 However, the prevalence of mild or moderate internet addiction in our study was 41.7%, which is higher than found in similar studies conducted in Pakistan (16.7%), 8 Qatar (17.6%), 7 Turkey (24.3%), 6 and Japan (38.2%), 13 but lower than reported by a Japanese study with undergraduates (48.5%). 9 In addition, the prevalence of more severe cases of internet addiction in this sample (7.5%) was higher than in the Brazilian general population (4.8%). 27 The sample under study is composed mainly of young individuals; a generation that has grown up in a world already fully connected by the internet, therefore constituting a subgroup with greater likelihood of internet addiction. 28 The prevalence of depressive symptoms observed in this investigation (18.7%) seems to agree with what would be expected for this population. A multicenter study with 12,677 Brazilian undergraduates from public and private institutions, using the same instrument, reported a similar prevalence, with 18.6% of participants having depressive symptoms. 29 However, this proportion was higher than among the Brazilian general population (14%). 12 Undergraduate students have being recognized as a subgroup with a high frequency of depressive symptoms, due to the stressful academic background, excessive workload, and developmental stage susceptibility. 10 A dose-response effect was observed between internet addiction and depressive symptoms. Other studies investigating internet addiction have also found an association with depressive symptoms. 6,7,13 A study with Japanese undergraduates also reported a dose-response relationship between these two variables, controlling for confounding variables. People with moderate and severe internet addiction were, respectively, 2.9 and 7.3 times more likely to have depressive symptoms compared to those without internet addiction. 9 In a longitudinal study, participants with higher internet addiction scores were more likely to develop mental health problems, such as depression. 30 One hypothesis to explain this result is a vicious cycle between internet addiction and depression. People may use the internet to reduce stress, increasing internet use and, consequently, decreasing social contact.
In turn, deterioration of interpersonal relationships may generate more stress, exacerbating depressive conditions, and making them seek even more relief from the internet. 9 One important result of our research was the mediating role of levels of physical activity on the association between moderate internet addiction (but not mild addiction) and depressive symptoms.
In other words, participants with a higher degree of internet addiction are more likely to report depressive symptoms, which was partly explained (10.7%) by lower levels of physical activity. No studies have been identified investigating physical activity as a potential mediator of this association. On the other hand, several studies have identified an association between internet addiction and low levels of physical activity, [6][7][8] as well as between physical activity and depression. 14,31,32 There is evidence to support this mechanism and its occurrence is both psychologically and biologically depression. This can occur by exchanging the negative reinforcement "stress relief through the internet" for another mechanism, "stress relief through physical activity," and by adding some positive reinforcement from physical activity (such as release of endorphins and serotonins and the feeling of performing an activity that is socially considered as healthy). Thus, the probability of eliminating the behavior of using the internet excessively is increased and depressive symptoms are reduced. 33 Lastly, this study must be interpreted within its limitations. First, its cross-sectional design does not allow the temporality of events to be observed. Second, physical activity assessment was based on self-report measures, which can be influenced by mood states, such as depression. Third, the course taken and the shift of study of the undergraduates may constitute possible sources of confounding of the results, but these data were not collected. However, this is an innovative study, since no prior research was found that in addition to investigating the relationship between internet addiction and depression has also identified a mediating role of physical activity in this association.

Conclusions
Therefore, we conclude that the prevalence of internet addiction and depressive symptoms in this sample was high, and that there is a dose-response relationship between them. The higher the levels of internet addiction, the greater the probability of depressive symptoms, even when controlling for potential confounders such as socioeconomic, demographic, and behavioral factors. Levels of physical activity mediated 10.7% of the association between moderate internet addiction and depressive symptoms, indicating that it may be an important therapeutic tool. A multiprofessional approach is recommended for treatment and monitoring of cases, considering the expertise required to prescribe and follow-up physical activities, psychotherapy, and/or pharmacotherapy.